GEOPHYSICAL EXAMINATION OF THE CHRISTIAN ARCHAEOLOGICAL SITE EMMAUS-NICOPOLIS (CENTRAL ISRAEL)

L. V. Eppelbaum a, *, S. E. Itkisb

aDept. of Geophysics and Planetary Sciences, Raymond and Beverly Sackler of Exact Sciences, Tel Aviv University, Ramat Aviv 69978, Tel Aviv, Israel - [email protected] bDept. of Geological and Environmental Sciences, Ben Gurion University of the Negev, Be’er-Sheva, Israel - [email protected]

KEY WORDS: Archaeology, Geophysics, Surveying, Application, Mathematics, Development, Interpretation, System

ABSTRACT:

Christian archaeological site Emmaus-Nicopolis is well known in the ancient and Biblical history. The site located halfway between Jerusalem and Tel Aviv, first built in the 5th century, over the site believed to be the place where Jesus appeared to two of his disciples after his resurrection. The Crusaders rebuilt it on a smaller scale in the 12th century. Two sites were examined by detailed magnetic investigations: (A) 25 x 40 m and (B) 10 x 24 m. Distance between the observation points was 1 meter, but not all points were accessible due to dense vegetation and rugged topography. Quantitative interpretation of magnetic anomalies was conducted using modern quantitative methods specially developed for complicated environments: oblique magnetization, rugged terrain relief and unknown level of the normal field. A distinct peculiarity of the survey was the fact that from these areas an upper part of soil (about two meters) containing modern contamination targets has been recently removed. A primary aim of this investigation was detection of buried ancient tunnels partially discovered at the eastern part of Emmaus-Nicoplis. However, performed survey allowing to revealing at least three high-intensive positive anomalies at the area A and one significant anomaly at the area B. Thus, all revealed anomalies (after removing 2m soil) must reflect some buried ancient remains. Determined depth of the upper edge of anomalous sources ranges from 0.7 to 1 m. Reliability of performed quantitative interpretation was successfully confirmed by 3-D modeling of magnetic field. The obtained results (they may have a great archaeological importance) were transmitted to archaeological group working at this site. The proposed ancient targets will be archaeologically inspected at the nearest time (apparently, until October 2003).

1. INTRODUCTION estimate the possible archaeological significance of the area under study. Rapid (first results may be obtained during a few The territory of Israel, in spite of comparatively small hours − several days) and reliable interpretation of magnetic dimensions (21,000 km2), is very attractive for archaeologists data should provide protection of archaeological remains from taking into account its dramatic ancient and Biblical history. It unpremeditated destruction. is undisputable fact that location of archaeological sites at Interpretation of magnetic surveys in Israel is complicated by a Israeli territory is the densest in the world (for instance, Meyer, strong oblique magnetization of the Earth’s magnetic field 1996; Reich, 1992). Christian remains consist of significant (about 42-44o). The multi-layered and variable structure of the part of the total number of discovered archaeological objects. upper part of the geological section (Dan, 1988; Rabikovitz, The Christian remains, according to the accomplished 1992) often does not allow calculating the level of the normal experience, occur in the subsurface layer at depth from 0 to 3 magnetic field within the studied sites. Noise caused by meters and often hold their initial correct (quasi-correct) industrial iron and iron-containing objects sometimes reaches geometrical form. Detailed magnetic survey is successfully high values. Rugged relief also disturbs the effect from the applied to searching and localization of the remains, as rapid, buried objects and complicates quantitative interpretation of effective and non-invasive tools for revealing a broad range of magnetic anomalies (Eppelbaum and Khesin, 2001). The various targets: buried walls, columns, foundations, complicated conditions of the survey require application of underground tunnels, chambers, water pipe systems and high sophisticated magnetic equipment, advanced methods of temperature features (Dalan, and Banerjee, 1996; Eppelbaum, qualitative and quantitative interpretation as well 3-D modeling 2000; Frese and Noble, 1984; Herwanger et al., 2000; of magnetic field. The developed methods (Eppelbaum et al., Weymouth, 1996). Geophysical surveys provide a ground plan 2000b; Khesin et al., 1996) allowed to eliminate noise, to reveal of cultural remains before excavations or may be even used archaeological remains and calculate their depth and size, and instead of excavations. Road and plant construction, selection to conduct an accurate 3-D modeling of magnetic fields. of areas for various engineering and agricultural aims are usually accompanied by detailed geophysical (first of all, Eppelbaum et al. (2003b) have shown an importance of correct magnetic) investigations. Such investigations should help mathematical formalization of geophysical/archaeological

* Corresponding author.

examination. The main role in the proposed algorithm plays a notion “information value”. Unfortunately, from the author’s experience follows that majority of geophysicists and archaeologists have troubles with accepting this approach.

Objects of archaeological study occur at a small depth and, consequently, the distance ∆x between profile observation points usually varies from 20 cm to one meter. Distance ∆y between profiles may not exceed ∆x by more than three times and ideally ∆y must equal to ∆x. Selection of a magnetic sensor level (practically it ranges in interval of 0.1 - 3 meters) depends on the concrete archaeological/geological situation.

M U M E NCE M L U C I E L A N T M U VE- D ; I ; E L Q AND G DE T ITIA - U OP N E N OF

P OF O 5 NG DE DE . Q F O IN S T O DEL THO

E I O 2 N; LI , N F E O TIO - ETIN ILITY S O D M M E 2 NG N O A R B G DE E

ARCHAEOLO- O M M - M CAT T O TI N OF O I I IN RP D O I F NG M TIM T CATI I EN RE CAL LOP AT I ELIA LOP E TE I CAL NAL DI YI Z L M I ES D I R R G I - AT RP L O VE P P RAP F IN P 3 VE P Z YS O AL TE AP DE M L H DE AP N P ALI TER I RE IN RE , S , S D S O S D H

OF F O

)

RAM

ON TI SOLU M E OBL R P E S R E V N T I AND O N G ) S RAM N O O T AND ETH N MET E I G S R T G O M HM M EN IO P M P R T T G IN D I H M O

C

T

P E

T LEC R

L ON I T LU SO BLEM O R P T C E R I E D R E O E ED P RI R S G ETIN V R P ELE ELO I R E S GO V P U AL (D E R Q TER REQUI AL D ( TE IN RE IN N S D

D

- N

L

N O I S I C PRE E O I NG I HY S R F GGE TI I D U S R OP O ELIEF I R D NDI

R E DEL O THEI

GE

CAL TH

S H O I Y T I L I B A H ST OF E OF F CO N S T M S N O O E M N O AND N G TI S M TI A AND T E

TIO

A

I T S

TIO N O I T I D N O C N O TI ZA I R LA O I P NG A T E M

A ETIN I

I

T M E U Q I L B O O T Y IT IL B A IC L P P TI A I L R T TI I I S LI B L RE S TIM RP E I I - LIM D S P B

ES RP

S I TE O NG

S N O I T I D N O C EF ELI S R I O

H

S IN TE

P D E G G RU O T Y T I L I AB C I L APP T TER O N I P RET IN ME

D S F IS ASS D S S S L L RVE S S D N O Y E U E C E I L RVED S O

O Figure 2. Exposed remains of walls in the central part of F E I E CT B I D

AL UR E O L RING

J F D AT A ARI RE AN E F Emmaus-Nicopolis site L B E Z C

OBS V T I O I A SI O C F

S L C

S O O I I Y I

S F ND DE H S

OF CAL PR I SUAL YS A E HY O

OP YS H VI L OP T GE ANAL

GE 2. BRIEF DESCRIPTION OF THE EMMAUS-

ANAL NICOPOLIS ARCHAEOLOGICAL SITE

S T L U S RE RK O W

- US O I V E PR E H T F O S I YS AL AN - - L CAL OF Christian archaeological site Emmaus-Nicopolis is well known I CA

ACE CAL

I I ) ON I T ZA I R LA O P E QU LI OB G N I D U L C N I ( T P ICAL CAL

S

ACE I

G

G in the ancient and Biblical history. The site (its fragment is S G N I H S I L P M CO AC A DAT CAL I S Y H P O R T E P P R L CAL G O S S EN O I OLOG Y O L O R E L H L M O DE O shown in Figure 2) located halfway between Jerusalem and Tel AND S G O E O P S P CAT G

AND

E

M L S ON TI A N ZO F E ELI R N I A R R AE TE OF CAT MO A E Aviv, first built in the 5th century, over the site believed to be S M NDI I L IA OF I ELO H B S NDI I LE V YS I IT O the place where Jesus appeared to two of his disciples after his ARCH E R

OB P IN N O I T A C I F I S AS CL D AN D ARC ARCHAEOLOG R

P S I YS AL AN A DAT CAL I G O L O E G A H ANAL ARC resurrection. The Crusaders rebuilt it on a smaller scale in the ANALYS 12th century (Mayer, 1996). Nicopolis is assumed in almost all Figure 1. Flow-chart of geophysical data processing and Christian Pilgrim texts from the 4th century onward. In 221 interpretation at archaeological sites (after Khesin et al., 1996, C.E. the Emperor Elagabalus gave Emmaus the title of city and with modifications) the name Nicopolis.

The primary aim of magnetic investigations at this site was According to our experience [Eppelbaum and Khesin, 2001; detecting underground tunnels (caves) partially investigated at Eppelbaum et al., 2000a,2000b,2001,2003a] the general scheme the eastern part of the area. However, purpose of our of geophysical data processing and interpretation at investigations was suddenly changed during the field archaeological sites may be composed using the following exploration. procedures presented in flow-chart (Figure 1).

3.2 Quantitative interpretation of magnetic anomalies 3. MAGNETIC EXAMINATION OF TWO SELECTED AREAS AT THE SITE OF EMMAUS-NICOPOLIS This stage involves application of methods for quantitative interpretation of magnetic anomalies for development of an 3.1 Methodology of field magnetic observations initial physical/archaeological model. The developed methods The magnetic survey at the site of Emmaus-Nicopolis have 0 4

been carried out using following equipment: , m • Proton Magnetometer MMP-203 (Era Assoc., Sankt- Petersburg, Russia), No. 206033 (it was used for field ance Dist

observations), 35 • Quantum Magnetometer MM-60 (Era Assoc., Sankt- Petersburg, Russia), No. 207001 (it was used for registering temporary magnetic variations at control point), 30 • Kappameter KT-5 (Scintrex, Canada) for magnetic susceptibility measurements.

Results of archaeological excavations and brief geological 25 examination of the areas under study indicate that we have here

two types of geological associations: soil (0 - 1.5 m) and Anomaly III F

underlying it limestone sometimes exposed at the earth’s 20 surface. A few tens performed measurements of magnetic D susceptibility indicate that limestone is practically non- E magnetic, and soil is characterized by magnetic susceptibility of B

60-100 x 10-5 SI. 15

I

C

y Detailed magnetic investigations were performed at two l Anomaly II

conjugated areas A and B (Figure 3). Unfortunately, small part a 10

m of area A and significant part of area B have been not covered o

by magnetic survey (see Figure 3) due to zone of very dense n A vegetation (area A) and zone of rugged topography (area B). 5 0 A

10 0

Area A 0 5

m 10 15 20 25 m , e c n a t s , Di e c n a t Figure 4. Map of the total magnetic field over area A at s

Di 20 Emmaus-Nicopolis Area B (improved modifications of characteristic point method and unsurveyed parts 30 tangent method are applicable in conditions of the rugged terrain topography, arbitrary direction of magnetization of the 0 1020304050 Distance, m objects and unknown level of the normal magnetic field (Khesin Figure 3. Location of two investigated areas et al., 1996). For quantitative interpretation of magnetic anomalies due to disturbing objects two geometric models were Magnetic sensor level, according to experience of previous utilized: thin bed (Figures 6, 8 and 9) and horizontal circular works (Eppelbaum et al., 2000a, 2000b, 2003), was selected as cylinder (Figure 7). It should be noted that anomaly I (Figure 6) 0.5 m above the earth’s surface. Temporary magnetic variations has a form close to ideal of theoretical anomaly due to a thin were removed using conventional scheme (Telford et al., 1998). bed. By its interpreting was used so-called Reford’s point Maps of the total magnetic fields over the areas A and B are (Reford and Sumner, 1964) showing a projection of the middle presented in Figures 4 and 5, respectively. Mean-square error of anomalous object to the earth’s surface. The disturbed upper for magnetic observations at the area A was 1.85 nT and at the part of anomaly II was smoothly reconstructed (dash line in area B – 0.9 nT. Figure 7) for the more convenience interpretation. Anomaly III The most important peculiarity of this survey is that from the is also disturbed by some nearly occurred magnetoactive areas under study has been recently removed an upper part of object(s). Anomaly IV of comparatively small intensity is soil (two meter thickness) containing all modern contamination. registered in the vicinity of the proposed continuation of Thus, we can propose that three high-intensive anomalies excavated underground tunnel. Determined depth of the upper observed at the area A (see Figure 4) and significant anomaly edge of the targets ranges from 0.7 to 1 m. revealed at the area B (see Figure 5) are associated with the buried ancient remains.

Distance, m 40 42 44 46 48 50 150 d1 8 S N

100 Anomaly II 10 Profile C - D a l s e T o

12 n 50 a n , d l d2 fie 14 ic t

e 0 n g ma

16 l ta o

T -50

m 18

-100 stance, d4 d3 i 20 D

0481216 22 Distance, m soil (I ≈ 50-80 mA/m) 0 H -1

m I ≈ 6000 mA/m -2 24 h, limestone (I ≈ 0) pt -3

De -4 -5 26 Anomaly 4 0481216 Figure 7. Quantitative interpretation of anomaly II (area A) 28

30

. G

Figure 5. Map of the total magnetic field over area B at

Emmaus-Nicopolis

300 d1 S N S N 50 d1 250

25 a

200 l s Anomaly I e T 0 o

a 150 Profile A - B l n s a e T n -25 ,

o 100 d n l Anomaly III a e n fi , 50 -50 Profile E - F Reford's c d d d 2 i 5 l

point t e fie

n -75

0 g ic t e n ma g -50 l -100 ta o ma l T

ta -100 -125 o T -150 -150 d4 d3

-200 0481216 Distance, m -250 d4 d3 soil (I ≈ 50-80 mA/m) 0 0 5 10 15 20 -1 I ≈ 4000 mA/m

Distance, m m , -2 soil (I ≈ 50-80 mA/m) h 0 t -3 limestone (I ≈ 0)

-1 I ≈ 8000 mA/m Dep -4 m -2 h, limestone (I ≈ 0) -5

pt -3 De -4 0 4 8 12 16 -5 Figure 8. Quantitative interpretation of anomaly III (area A) 0 5 10 15 20 Figure 6. Quantitative interpretation of anomaly I (area A)

20 d 300 S 1 N S N

200 a l a l s s

Anomaly IV e e T

T 0 Profile G - H

no 100 no na na d 2 d, l d, l e i e i

f 0 f c i c t i e t e

-20 n n g g a

a -100 m l m a l Graphs of the total t a o t magnetic field: o T

T -200 Observed -40 Computed d4 d3 -300

04812 0 5 10 15 20 Distance, m Distance, m 0 soil (I ≈ 50-80 mA/m) 0 I ≈ 1300 mA/m I = 10000 mA/m -1 m -1 m

-2 h, h, pt pt -3 limestone (I ≈ 0) e

e -2 D

D -4 anomalous object -5 -3 04812 0 5 10 15 20 Figure 9. Quantitative interpretation of anomaly IV (area B) Figure 10. Results of 3-D modelling of magnetic field using

GSFC-1 program. Arrow indicates the direction of magnetic

vector 3.3 3-D modelling of magnetic anomalies

The GSFC-1 (Geological Space Field Calculation) program Finally, Figure 11 shows an image of the examined area B at was developed for solving a direct 3-D gravity and magnetic Emmaus-Nicopolis showing the projection of the upper edge of prospecting problem under complicated geological conditions anomalous body to the earth’s surface. On the basis of (Khesin et al., 1996; Eppelbaum, 2003). This program has been integrated analysis of geophysical and archaeological data we designed for computing the field of ∆g (Bouguer, free-air or may suggest that this anomaly (see Figure 9) may be produced observed value anomalies), ∆Z, ∆X, ∆Y, ∆T, as well as second by some archaeological remain(s) containing in underground derivatives of the gravitational potential under conditions of tunnel. rugged relief and inclined magnetization. The geological space can be approximated by (1) three-dimensional, (2) semi-infinite 4. Conclusions bodies and (3) those infinite along the strike (closed, L.H. non- closed, R.H. non-closed and open). Geological bodies are We can conclude that four analyzed magnetic anomalies with a approximated by horizontal polygonal prisms. high probability can correspond to buried archaeological The basic algorithm realized in the GSFC program is the remains. Despite of the fact that the recognized anomalies at the solution of the direct 3-D problem of gravimetric and magnetic area A have more high intensity than the single anomaly at the prospecting for horizontal polygonal prism limited in the strike area B, we propose that the last anomaly may have more direction. In the presented algorithm integration over a volume important archaeological significance and it should be is realized on the surface limiting the anomalous body. excavated for the first time.

Results of 3-D modelling of magnetic field produced by the Preferably, three magnetic anomalies displayed at the area A, buried target at area B have shown in Figure 10. must be examined also by conventional metal detector equipment. As initial model for the computing, the data obtained at the previous stage of quantitative interpretation (see Figure 6), were The areas of geophysical examination must be extended to the utilized. Figure 10 illustrates that observed and computed west where also archaeological remains may be found. graphs gave an excellent agreement. Thus, good coinciding the observed and computed graphs proves the reliability of performed quantitative interpretation. Similar results were References obtained and for anomalies II - IV. Dalan, R.A. and Banerjee, S.K., 1996. Soil magnetism, an approach for examining archaeological landscapes. Geophysical Research Letters, 23 (2), pp. 185-188.

Dan, J., 1988. The soils of the land of Israel, In: The Weymouth, J.W., 1996. Digs without digging: exploring zoogeography of Israel, Y. Yom-Tov & E. Tchernov (Eds.), W. archaeological sites with geophysical techniques. Geotimes, Junk Publishers, Dordrecht, pp. 95-128. Nov., pp. 16-19. Eppelbaum, L.V., 2000. Applicability of geophysical methods

for localization of archaeological targets: An introduction. lief Geoinformatics, 11(1), pp. 19-28. f Eppelbaum, L.V., 2003. Introduction to potential geophysical gged re ne o u fields. Handmanual for students, Tel Aviv University. zo r

Eppelbaum, L.V., Itkis, S.E. and Petrov, A.V., 2000a. Physics B and archaeology: magnetic field as a reliable tool for searching ancient remains in Israel. Scientific Israel, No.2, pp. 68-78. Area Eppelbaum, L.V., Itkis, S.E. and Khesin, B.E., 2000b. Optimization of magnetic investigations in the archaeological y IV

sites in Israel. Spec. Issue of Prospezioni Archeologiche, al

“Filtering, Modelling and Interpretation of Geophysical Fields m o at Archaeological Objects”, pp. 65-92. n A Eppelbaum, L.V. and Khesin, B.E., 2001. Disturbing Factors in Geophysical Investigations at Archaeological Sites and Ways of Their Elimination. Trans. of the IV Conf. on Archaeological Prospection, Vienna, Austria, 99-101. Eppelbaum, L.V., Khesin, B.E. and Itkis, S.E., 2001. Prompt magnetic investigations of archaeological remains in areas of infrastructure development: Israeli experience. Archaeological Prospection, 8(3), 163-185. Eppelbaum, L., Ben-Avraham, Z. and Itkis, S., 2003a. Ancient Roman Remains in Israel provide a challenge for physical- archaeological modelling techniques. First Break, 21 (2), pp. 51-61.

Eppelbaum, L., Eppelbaum V. and Ben-Avraham, Z., 2003b. Formalization and estimation of integrated geological investigations: Informational Approach. Geoinformatics, 14, No.3, pp. 233-249. Herwanger, J., Maurer, H., Green, A.G., and J. Leckebusch, 2000. 3-D inversions of magnetic gradiometer data in archaeological prospecting: possibilities and limitations. Geophysics, 65, No. 3, pp. 849-860. Khesin, B., Alexeyev, V. and Eppelbaum, L., 1996. Interpretation of Geophysical Fields in Complicated Environments. Kluwer Acad. Publisher, Ser.: Modern Geoph. Approaches, Dordrecht - London - Boston. Meyer, E.M. (Ed.), 1996. The Oxford Encyclopedia of Archaeology in the Near East. 5 vols., Oxford University Press, Oxford. Figure 11. Photography of area B with projection of the Rabikovitz, S. 1992. The Soils of Israel: Formation, Nature, proposed anomaly source to the earth’s surface and Properties. Hakibbuts Hamehuchad Publishing House, Israel (in Hebrew), with 1:250,000 soil map of Israel (in Hebrew and English). Acknowledgements Reford, M.S. and Sumner, I.S., 1964. Aeromagnetics. The authors are grateful to Dr. K.-H. Fleckenstein and Mrs. L. Geophysics, 29, No. 4, pp. 482-516. Fleckenstein – working archaeological group at Emmaus- Reich, R., 1992. Architecture of ancient Israel. Israel Nicopolis site – for very useful communications. Exploration Society, Jerusalem. Telford, W.M., Geldart, L.R. and Sheriff, R.E., 1998. Applied Geophysics. 5th ed. Cambridge University Press, Cambridge. von Frese, R.R.B. and Noble, V.E., 1984. Magnetometry for archaeological exploration of historical sites. Historical Archaeology, 18 (2), pp. 38-53.